Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
About this item
Full title
Author / Creator
Publisher
Ithaca: Cornell University Library, arXiv.org
Journal title
Language
English
Formats
Publication information
Publisher
Ithaca: Cornell University Library, arXiv.org
Subjects
More information
Scope and Contents
Contents
We study the training process of Deep Neural Networks (DNNs) from the Fourier analysis perspective. We demonstrate a very universal Frequency Principle (F-Principle) -- DNNs often fit target functions from low to high frequencies -- on high-dimensional benchmark datasets such as MNIST/CIFAR10 and deep neural networks such as VGG16. This F-Principle...
Alternative Titles
Full title
Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks
Authors, Artists and Contributors
Author / Creator
Identifiers
Primary Identifiers
Record Identifier
TN_cdi_proquest_journals_2170078314
Permalink
https://devfeature-collection.sl.nsw.gov.au/record/TN_cdi_proquest_journals_2170078314
Other Identifiers
E-ISSN
2331-8422
DOI
10.48550/arxiv.1901.06523
